Upgrade cassandra image to 4.1.3 in docker compose files

[taylanisikdemir]

What changed?
Docker compose files were referring to an older version (3.11) of Cassandra. Updated them with current GA version 4.1.3.

ubercadence/server:master-auto-setup container's initialization script waits on Cassandra to be up and corresponding wait command is updated to use --cqlversion=3.4.6 (previously 3.4.4).

Misc change: I was also interested in to see Cassandra's full query logs. Added docs/howtos/cassandra-fql.md to explain the steps on how to check FQL.

Why?
To keep things up to date. Assuming Cassandra 4+ is more widely used.

How did you test it?

• Locally run make test and bench tests
• Tried a few of the docker-compose-*.yml files

Potential risks
Users who depends on ubercadence/server:master-auto-setup should pull new image before running docker-compose. Hopefully this is only used for development/discovery use cases. Left a note in changelog for this.

[taylanisikdemir]

avoiding usage of port 9000 because it's a popular port and high chance of conflict. (I encountered this issue)

[SaurabhGoyal]

This is causing versioning issues when using docker-compose.yml to spin up the container -

cadence_1        | 
cadence_1        | Warning: Using a password on the command line interface can be insecure.
cadence_1        | Recommendation: use the credentials file to securely provide the password.
cadence_1        | 
cadence_1        | Traceback (most recent call last):
cadence_1        |   File "/usr/bin/cqlsh", line 8, in <module>
cadence_1        |     sys.exit(main())
cadence_1        |   File "/usr/lib/python3.9/site-packages/cqlsh/__main__.py", line 7, in main
cadence_1        |     cqlsh_main(*read_options(sys.argv[1:], os.environ))
cadence_1        |   File "/usr/lib/python3.9/site-packages/cqlsh/cqlsh.py", line 2329, in main
cadence_1        |     shell = Shell(hostname,
cadence_1        |   File "/usr/lib/python3.9/site-packages/cqlsh/cqlsh.py", line 476, in __init__
cadence_1        |     load_balancing_policy=WhiteListRoundRobinPolicy([self.hostname]),
cadence_1        |   File "/usr/lib/python3.9/site-packages/cassandra/policies.py", line 425, in __init__
cadence_1        |     self._allowed_hosts_resolved = [endpoint[4][0] for a in self._allowed_hosts
cadence_1        |   File "/usr/lib/python3.9/site-packages/cassandra/policies.py", line 426, in <listcomp>
cadence_1        |     for endpoint in socket.getaddrinfo(a, None, socket.AF_UNSPEC, socket.SOCK_STREAM)]
cadence_1        |   File "/usr/lib/python3.9/socket.py", line 954, in getaddrinfo
cadence_1        |     for res in _socket.getaddrinfo(host, port, family, type, proto, flags):
cadence_1        | socket.gaierror: [Errno -2] Name does not resolve
cadence_1        | + echo 'waiting for cassandra to start up'
cadence_1        | + sleep 1
cadence_1        | waiting for cassandra to start up
cadence_1        | + cqlsh -u cassandra -p cassandra --cqlversion=3.4.4 --protocol-version=4 cassandra

[taylanisikdemir]

This is causing versioning issues when using docker-compose.yml to spin up the container -

cadence_1        | 
cadence_1        | Warning: Using a password on the command line interface can be insecure.
cadence_1        | Recommendation: use the credentials file to securely provide the password.
cadence_1        | 
cadence_1        | Traceback (most recent call last):
cadence_1        |   File "/usr/bin/cqlsh", line 8, in <module>
cadence_1        |     sys.exit(main())
cadence_1        |   File "/usr/lib/python3.9/site-packages/cqlsh/__main__.py", line 7, in main
cadence_1        |     cqlsh_main(*read_options(sys.argv[1:], os.environ))
cadence_1        |   File "/usr/lib/python3.9/site-packages/cqlsh/cqlsh.py", line 2329, in main
cadence_1        |     shell = Shell(hostname,
cadence_1        |   File "/usr/lib/python3.9/site-packages/cqlsh/cqlsh.py", line 476, in __init__
cadence_1        |     load_balancing_policy=WhiteListRoundRobinPolicy([self.hostname]),
cadence_1        |   File "/usr/lib/python3.9/site-packages/cassandra/policies.py", line 425, in __init__
cadence_1        |     self._allowed_hosts_resolved = [endpoint[4][0] for a in self._allowed_hosts
cadence_1        |   File "/usr/lib/python3.9/site-packages/cassandra/policies.py", line 426, in <listcomp>
cadence_1        |     for endpoint in socket.getaddrinfo(a, None, socket.AF_UNSPEC, socket.SOCK_STREAM)]
cadence_1        |   File "/usr/lib/python3.9/socket.py", line 954, in getaddrinfo
cadence_1        |     for res in _socket.getaddrinfo(host, port, family, type, proto, flags):
cadence_1        | socket.gaierror: [Errno -2] Name does not resolve
cadence_1        | + echo 'waiting for cassandra to start up'
cadence_1        | + sleep 1
cadence_1        | waiting for cassandra to start up
cadence_1        | + cqlsh -u cassandra -p cassandra --cqlversion=3.4.4 --protocol-version=4 cassandra

You might have a previous sha of ubercadence/server:master-auto-setup on your machine. Try this to start fresh and let me know if it works

docker-compose stop
docker-compose rm -f
docker-compose pull   
docker-compose up